►
From YouTube: 2021 01 19 Database Team Weekly
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Welcome
to
this
version
of
the
database
team
weekly
meeting
is
january.
19Th
a
couple
topics
right
off
the
top-
won't
take
too
much
time.
So
it's
okay,
our
time,
I've
put
the
links
to
current
okr
and
draft
for
upcoming
okr.
Take
a
look
at
your
comments
and
then
jump
right
into
the
infradev
review
for
any
incoming
issues,
and
I
actually
did
this
yesterday.
There
were
no
new
issues
yesterday,
I
haven't
checked
yet
this
morning.
A
B
No,
I
just
would
like
to
review
some
issues,
one
of
them
I
added
some
marginalia
for
with
the
artist
to
have
a
look
and
proceed
with
the
teams
and
so
on.
If
because
I
remember,
we
did
some
raw
analysis
with
nikolai
and
we
collected
some
samples.
This
is
one
thing
and
the
other
issues
I
think
they're
assigned
to
other
teams,
but
we
didn't
have
any
update
on
them.
No,
but
there
is
no
new
issues
at
the
end.
Okay,.
A
B
Yeah
it
was
this
one
of
them,
the
other
one.
I
did
a
comment
to
you
a
few
hours
ago
minutes
ago,
janice
that
I
collected
for
the
name
space
square.
If
I'm
not
wrong
so
marginalia
there,
if
you
can
have
a
look
just
for
us
to
iterate,
would
be
awesome.
Okay,
thank
you
just
that's
it
from
my
side,
craig.
Thank
you.
A
Okay,
is
this
one
still
like
on
fire,
or
is
it
okay
that
it's
scheduled
for
13.9
like
do?
We
need
to
get
some
more
urgency
with
the
access
team
and
liam
to
try
and
get
it
pushed
to
the
top
of
the
priority
queue.
B
B
Something
and
we
already
had
applied
the
change
to
reduce
the
select
once
that's
great
yesterday-
was
public
holiday
in
us.
So
I
understand
that
we
had
lower
traffic
today.
I
think
the
next
hour
I
will
see
if
there
is
some
spike,
but
will
be
great
if
you
can
do
some
work
around
in
a
couple
of
issues
or
these
heavy
metals
that
we
had.
That
was
like
the
top
consumers
in
total
time
for
us
to
try
to
mitigate
the
whole
solution.
The
whole
problem
itself,
no
yep.
A
C
C
To
jose
so
the
master
on
gitlab.com
has
changed.
Is
this
related
to
an
incident
or.
B
Yes,
unfortunately,
sunday,
we
had
like
a
maintenance
to
expand
the
disks,
because
we
were
running
out
of
space.
We
have
our
databases
around
nine
terabytes
and
we're
in
a
disk
of
10.
and
what
happened.
Unfortunately,
one
of
our
colleagues
shot
or
started
a
snapshot
from
the
disks
trying
to
be
safe,
a
sorry
mentality,
and
unfortunately,
this
triggered
a
failure
like
we
are
still
investigating
this.
But
this
was
the
reason
that
we
had
a
failover
that
we
changed
the
primary.
It
was
not
expected.
Okay,.
A
It's
not
good
that
we're
using
that
much
space,
but
it's
good
that
we've
got
more
space,
more
overhead,
okay,
yeah
I'll
chat
with
our
product
folks
about
getting
some
prioritization
on
this
issue
later
on.
A
All
right,
if
there's
nothing
else
jose,
you
can
drop
off.
A
Everyone
all
right,
so
I'm
just
taking
a
look
at
our
board.
We
still
have
a
bunch
in
13
8.
I
don't
know
it's
been
updated.
Let
me
refresh
real
quick,
so
andrea's,
obviously
out
so
those
will
carry
over
if
he
doesn't
move
them
the
bottle
carry
them
over
patrick.
You
have
a
couple
that
are
in
here
so
properly.
Sync
data.
C
And
they
have
already
by
so
there
are
the
13.8
on
top,
then
there
are
the
13.9
and
then
everything
else.
D
A
I
think
this
is
part
of
the
the
growing
pain
of
changing
process
which
I'm
totally
on
board
for
it's.
The
the
sedu
ratio
clash
that
we're
gonna
have
where
I
think,
where
we're
coming
up
against
the
end
of
a
milestone,
and
things
are
still
in
that
milestone,
figuring
out
what
we
need
to
do
with
those.
So
I'm
fine
with
going
this
route,
I
think
there's
another
one.
Yeah.
D
My
suggestion
what
I
usually
do
with
the
geo
team
is,
we
just
start
at
the
right
and
see
what's
done,
and
then
we
sort
of
make
our
way
to
the
left
and
see
if
there
are
things
that
need
special
attention
depending
on
where
they
actually
are,
and
then
you
can
that's.
That's
always
that's
what
we
do.
D
E
D
A
D
Cool,
let
me
quickly.
D
Refresh,
can
everybody
see
the
board
cool?
Okay
item
number
one
13.9
planning.
I
think
that
what
got
done
you
recorded
the
kickoff
call
as
well
ganes,
that's
right!
Yes,
then,
the
next
one
here.
D
E
D
I
usually
do,
and
then
I
can
also
do
it
later
on.
Is
I
unlabel
them,
so
they
will
no
longer
actually
appear
here
in
the
column
in
that
way
to.
A
D
Perfect,
okay,
see
I
check
for
mr
patrick.
E
Yeah
that
one's
done
yeah
there
was
a
follow-up
created,
but
the
work
captured
here
is
finished.
B
C
D
Okay,
so
started
to
work
on
this.
C
A
D
A
E
D
Okay,
and
so
this
was
labeled
to
be
released
in
13.8,
I
think
that
chip
has
sailed,
correct,
yeah,
yeah,
okay,
I'll
just
update
the
milestone
here
then.
D
D
C
B
C
Check
it
afterwards,
maybe
we
added
it
while
testing.
C
We
have
backlogged
it
so
just
yeah,
I
I
will
address
it
most.
Probably
we
will
have
to
remove
the
database
active.
D
Yeah,
that's
good
okay
in
dev,
but
the
big
end
conversion.
C
Yeah
this
one
is
the
one
where
we
everything
is
ready,
but
we
are
going
to
need
six
months
for
it
to
learn.
So
we
have
to
run
some
tests
or
figure
out
a
way
to
have
better
statistics.
C
Most
probably
we
will
have
to
split
the
mars
and
one
more
merge
request
there
so
that
we
can
gather
statistics
and
start
it
off
most.
Probably
this
should
not
take
six
months.
It
should
take
two
months
but
yeah
we
have
to
to
check.
D
Okay,
but
this
so
this
is
actually
in
dev.
It
just
requires
more
mrs
to
be
raised
rather
than
on
review.
Okay,.
C
Yeah,
I
will
need
help
and
most
probably
we
need
help
for
the
for
the
follow-up
to
this
one
that
will,
we
will
need
to
add
the
migration
helpers
for
finalizing.
So
maybe
I
will
continue
with
this
one,
and
then
I
have
already
asked
whether
patrick
or
andreas
can
help
me
with.
A
D
This
is
this
is
by
the
way,
also
what
we
do
in
the
scheduling
call
in
in
geo.
We
often
when
these
discussions
happen.
We
just
we
just
add
them
here
as
as
commons
into
the
into
the
issue,
so
they
don't
get
get
lost
exit
base.
A
Based
on
urgency
and
timing
like
this
should
be
the
top
thing
that
either
pat
or
andreas
jump
on
as
time
as
as
they
have
free
time
and
are
able
to
help
yeah.
Because
there
is.
D
I
think
that's
that's
a
good
call.
Okay
onwards
index
renaming.
E
Yeah
this
one
I
think
andreas
had
mostly
complete.
There
was
a
one
or
two
minor
issues
pointed
out
during
the
review
that
he
just
hasn't
had
time
to
address
yet,
but
it
should
be
good
for
is
this
in
review
or
in
depth.
It
was
yeah
and
I
think
he
had.
He
maybe
put
it
back
in
depth
because
he
had
to
make
a
few
changes,
but
obviously
it
slipped
13.8,
but
it
shouldn't
go
in
13.9.
B
C
C
C
D
Okay,
then
ready
for
development,
so
my
understanding
for
ready
for
development
is
that
no
work
has
been
done
on
on
those
yet
right.
It's
just
people
will
will
pick
up
this
work.
D
E
There
was
a
review
that
I
had
and
I
determined
that
there
was
a
bug
with
the
how
the
helper
works
currently,
and
so
it's
holding
up
this
merge
requested
sitting
there,
and
so
I
thought
it
would
be
relatively
simple
to
fix
it
and
get
it
in
turns
out.
That's
it's
not
going
to
be
as
simple
as
I
thought
and
basically
have
to
rewrite
or
create
a
new
helper.
E
D
A
D
D
Okay,
so
looking
at
those
here,
the
next
ones
I
want-
maybe
we
can
go
on
through
them.
High
level
are
those
the
deliverable
items
for
13.8
in.
C
D
C
C
Yeah
from
determining
the
first
two
are
in
order
to
finish
with
the
audit
events.
Partitioning
is
just
running
some
tests
and
then
dropping
the
table,
and
they
should
be
deliverables.
This
is
the
important
one
for
what
will
allow
us
to
finalize
the
migration
on
very
stable.
This
will
be
from
my
perspective,
and
then
there
are.
C
D
Okay-
and
these
here
are
all
of
the
operational,
the
the
tail
end
of
things
is
that
correct,
yeah,
okay,
so
my
concern
is,
I
think
we
should
quickly
go
back
to
the
milestone
board.
I
just
want
to
make
sure
we've
we've
accounted
for
everything,
and
this
is
the
you
know.
The
transition
bit
as
as
craig
pointed
out
earlier
are
all
of
those
items
in
here
the
same
as
in
our
milestone
board,
which
I
need
to
pull
up.
C
D
C
Yeah
everything
is
ready
and
following.
D
This
is
okay
cool.
So
a
couple
of
observations,
just
from
my
experience
before
we
talk
about
that
these
things
here.
This
is
a
long
list
right,
it'll
take
us
a
while
to
actually
like
go
through
this,
so
I'll
leave
it
like.
Maybe
we
can
do
this
async,
but
these
should
be
ordered
in
in
rough
priority.
D
I
personally
prefer
having
like
fewer
than
10
items
in
here,
because
I
think
everything
else
is
a
little
bit
of
a
lie
right
as
in
we're
going
to
like
there's
tons
of
stuff.
That's
ready
for
development.
That's
true,
but
you
know
we
can
sort
of
refill
the
pile
you
know
continuously
over
the
milestone,
rather
than
having
them
all
all
in
here,
because
it
gets
a
little
bit
unclear
what
to
actually
pick
and
choose
right.
D
A
C
This
is,
I
think
that
we
would
like
to
to
work
on
at
some
point,
but.
E
Backlog
yeah
that
last
one
is
a
follow-up
to
the
one
that
we
just
closed
recently.
So
we
added
the
ci
immigration
job
and
we
just
have
to
basically
flip
it
to
to
be
on
essentially.
D
D
So
my
like
again,
this
is
a
discussion,
but
this
is
what
what
I
do.
It's
the
ready
for
development
should
be
ideally
stack
ranked
by
priority.
You
know
the
things
that
are
most
important
are
at
the
top
and
then,
let's
say
patrick
picks,
something
picks
the
next
thing,
and
I
mean
for
you.
This
would
be
this
one
here,
right
and
jan
is,
if
you
had
capacity,
which
you
don't
right,
the
next
thing
you
would
pick
is
the
partitioning
audit
right.
D
So
the
lower
things
here
are
the
like
less
the
priority
right,
that's
kind
of
how
it
is.
As
I
said,
I
like
to
not
keep
too
many
in
here,
because
otherwise
you
it's
a
little
bit
like
coffee,
grinds
in
a
cup
right,
you,
you
dump
things
to
the
bottom
and
they
never
quite
make
it,
and
so
I
think
the
activity
here
is
to
make
sure
that
every
week
you
re-evaluate
if
these
are
still
the
most
important
things
to
actually
work
on
right
and
not
pull
in
too
many
new
things.
D
There
are
two
more
right
which
I'm
adding
now
I
don't
keep
the
entire
backlog
in
the
open
column,
and
this
is
how
I
keep
the
the
board
relatively
clean,
because
the
for
me,
the
importance
in
this
board
is
that
it
is
like
manageable
to
actually
talk
about
it
within,
let's
say,
20
minutes
or
so
or
15
minutes,
rather
than
having
to
go
through.
Like
tons
of
issues
that
you
know,
you
will
not
complete
within
the
next
few
weeks
where
it's
like.
D
If
we
didn't
do
anything,
I
asked
myself
it's
like:
when
will
we
actually
get
to
this
item
right?
Is
it
going
to
be
the
end
of
13.9
right
or
is
it
going
to
be
in
a
week?
And
so
this
is,
this
is
sort
of
my
thought
process.
B
C
In
the
open
column
ready
to
be
added
to
for
ready
for
development,
I
don't
think
that
we
can
make
them
in
30.9.
I
don't
know
if
everyone
else
agrees
so
yeah.
C
D
We
actually
want
to
move
them
into
your
planning
board.
Okay
right,
that's
maybe
the
I
should
have
done
that.
Otherwise,.
A
D
Yeah,
I
well
for
technical
reasons.
I
fail
to
do
this,
but
we
can
still
agree
on.
We
can
move
these
back
into
the
planning
board
right
and
then
decide
when
to
actually
do
them,
and
I
would
argue
these
four
here
can
be
the
same,
except
for
maybe
the
one
that
patrick
pointed
out,
because
he
said
that
as
a
follow-up
to
something
that
was
done
recently.
So
it
may
make
sense
to
do
it
as
well.
C
A
D
C
Them
and
and
add
them,
as
announced
at
the
validation,
the
scheduling,
workflow.
D
Scheduling
so
actually,
just
as
a
little
like
this
is
what
I
have
for
for
geo
right.
I
have
all
of
my
short
codes
for
labels
set
up
so.
D
A
E
D
E
D
A
C
So
for
the
migration
helpers
on
the
left,
I'm
unassigned
so
hopefully
patrick
can
pick
it
on
the
other
one,
the
migration.
I
will
try
to
follow
up
with
that
and
if
I
cannot,
I
will
politely
ask
her
but
or
someone
but
to
help.
D
Me,
okay,
okay,
but
that's
kind
of
it
I'm
going
to
like
stop
sharing
now
my
laptop
is
about
to
explode,
but
so
the
idea
here
is
that
you
know.
Usually
this
is
a
bit
faster.
We
do
this
for
the
first
time
right,
but
that
you
just
go
to
the
billboard
and
then
pick
the
next
thing
from
the
like.
D
What's
what's
in
the
ready
for
development
column
and
can
like
start
working
on
that
and
the
the
discussion
to
have
for
all
of
us-
but
mainly
you
know,
for
yanis
going
forward-
is
to
make
sure
that
the
items
for
the
prioritized
epics
that
we
want
to
work
on.
So
the
main
theme
for
for
this
milestone
are
ranked
high
enough
so
that
they
actually
get
delivered
and
a
risk
here
is
always
that
you
fill
your
workflow
like
you're,
ready
for
development
stuff.
D
So
full
you
know
that
the
bottom
is
never
actually
going
to
be
worked
on
because
they
you
know
they
don't
they
never
make
it
in
priority
and
that's
something
we
have
in
geo
pretty
frequently
and
I
just
remove
them
because
otherwise,
your
you
know
you're
essentially
signaling.
We
are
going
to
get
to
it,
but
if
you've
tried
to
get
to
it
for
like
two
months
and
it
hasn't
happened,
you
know
it
doesn't
really
reflect
reality.
D
A
So
when
does
the
triage
portion
take
place
right
because
we're
going
to
be
different
than
future
teams
and
that
we
get
a
fairly
high
volume
of
issues
coming
in
that
need
to
be
looked
at,
and
it's
going
to
shuffle
our
priorities
on
a
weekly
basis
right.
So
how
do
we
want
to
incorporate
that
step
into
our
processes?.
D
So
my
my
initial
suggestion-
and
I
I
don't
I
don't
have
that
much
experience
with
it-
is
that
we
do
this
in
geo
now
either
there's
a
rotation
for
triaging,
so
somebody
is
going
to
do
all
of
the
triaging
for
a
milestone
so
that
you
know
you
can
ignore
it.
You
know
for
a
month
and
somebody
just
does
it,
but
I
think,
given
that
you
have
a
higher
volume
right,
I
think
what
needs
to
happen
is
by
the
time
you
have
the
database
meeting
where
you're
going
to
walk
the
the
board
right.
D
The
prioritized
and
triaged
issues
should
be
ready
right.
You
need
to
be
able
to
say
hey,
we
have
these
three
things
that
came
in
you
know,
maybe
put
them
in
the
open
column
and
say
like
these
are
high
priority
things.
D
We
need
to
do
them
right,
but
if
they,
if
they
that's
kind
of
probably
the
the
best
we
can
do
honestly,
you
know
if,
if
this
is
a
little
bit
established
right,
you
can
also
pull
them
into
ready
for
development
immediately
after
triaging
right,
because
if,
if
it's
understood
then
the
most
important
thing
is
at
the
top
of
ready
for
development
and
janice
is
on
on
triaging
duty
and
finds
a
p1
s1
tomorrow
right,
then
you
just
pull
it
into
ready
for
development,
and
that's
the
next
thing
that
people
are
going
to
do,
which
is
essentially
the
the
definition
of
of
that
severity
right.
D
A
A
D
A
That's
a
good
way
to
get
people
to
talk
about
it.
That's
true!
All
right,
then
we
can
put
that
at
the
in
the
agenda
too,
like
whoever
was
on
triage
last
week,
bring
up
any
items
or
make
sure
they're
represented
on
the
board
and
hand
off
to
the
next
person.
So
we
can
just
go
in
alphabetical
order,
then
to
make
it
simple,
so
pat
giannis
and
andreas,
so
we
almost
have
next
week.
Andrea,
says
the
week
after
that
makes
it
easier
to
figure
out.
D
Yeah,
that's
cool,
no,
and,
and
also
the
again
like
the
way.
The
mindset
I'm
trying
to
sort
of
get
across
is
just
look
at
the
billboard
or
that
should
always
reflect
the
reality
of
what
is
most
important
and
that's
the
the
beauty
of
it
right
at
some
point
you
can
you
know
during
the
week.
If
you
need
something
to
do,
you
should
be
able
to
just
go
there
and
pick.
A
That's
it
for
topics
on
the
agenda.
Anything
else
folks
wanted
to
cover
today.